Non-contact imaging of peripheral hemodynamics during cognitive and psychological stressors

Figures

• 

  Figure 1. Peripheral hemodynamics contain rich information about physiological state and health. We present analysis of non-contact camera-based measurements of pulse rate variability, blood volume pulse amplitude and vasomotion signals during three experiments eliciting stress.

• 

  Figure 2. (i) Vasomotion, (ii) peripheral pulse amplitude, (iii) PRV LF normalized power, (iv) PRV HF normalized power and (v) PRV LF/HF power ratio across all three tasks for the rest and stress conditions. The error bars reflect the standard errors.

• 

  Figure 3. Individual-level measures of: (top row) vasomotion power, (middle row) peripheral blood volume pulse power, (bottom row) PRV LF/HF ratio. For each individual we calculated a single measure per condition.

• 

  Figure 4. Auditory stimulation example. The total blood concentration (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {C}_{tb}$$\end{document}Ctb), vasomotion, and detrended peripheral pulse amplitude plots. A magnified scan line from the video is shown at the top. A, B Normalized total blood concentration at rest and under stress respectively. C, D Normalized vasomotion at rest and under stress respectively, E, F Normalized detrended peripheral pulse amplitude at rest and under stress respectively.

• 

  Figure 5. Stroop example. The total blood concentration (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {C}_{tb}$$\end{document}Ctb), vasomotion, and detrended peripheral pulse amplitude plots. A magnified scan line from the video is shown at the top. A, B Normalized total blood concentration at rest and under stress respectively, C, D Normalized vasomotion at rest and under stress respectively, E, F Normalized detrended peripheral pulse amplitude at rest and under stress respectively.

• 

  Figure 6. MAT example. The total blood concentration (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {C}_{tb}$$\end{document}Ctb), vasomotion, and detrended peripheral pulse amplitude plots. A magnified scan line from the video is shown at the top. A, B Normalized total blood concentration at rest and under stress respectively. C, D Normalized vasomotion at rest and under stress respectively. E, F Normalized detrended peripheral pulse amplitude at rest and under stress respectively.

• 

  Figure 7. To evaluate peripheral hemodynamics we calculate pulse rate variability, pulse amplitude and vasomotion measures. We capture videos from digital cameras, segment a skin region (either on the face or on the hand) and calculate spatial averages of the color channel signals within the region of interest. We then recover the total blood concentration using a calibrated model-based approach. Finally we extract frequency domain measures of the physiological parameters. The face images used are examples provided with consent to be published in an online open-access publication.

• 

  Figure 8. Physiological measurements using a camera (thick black lines) in comparison with a finger BVP contact sensor (thin gray lines). a PPG waveforms normalized from 0 to 1. b Detrended BVP waveform. c Inter-beat intervals (IBIs) formed by extracting the peaks from the BVP waveforms. d Normalized Lomb periodogram of the detrended IBIs. Notice in particular how the PPG waveform dynamics, amplitudes and inter-beat intervals are closely aligned.

• 

  Figure 9. Experimental set-ups. Experimental set-up a was used for data collection in the auditory stimulation and stroop stimulation experiments. Experimental set-up b was used for data collection in the MAT stimulation experiment.